Materials and methods
Modern oak samples
Modern oak pollen samples of 23 species were collected from herbarium
specimens at the University and Jepson Herbaria of the University of
California at Berkeley, from living plants at University of California
Botanical Garden at Berkeley, and from living plants at Tilden Regional
Park, Regional Parks Botanic Garden (Table 2).
Floral materials were removed from specimens and branches by cleaned
(sterile) forceps and stored in 95 % EOTH in 5 ml bottles. For pollen
extraction, we first transferred samples, smaller than 125 microns, from
5 ml bottles to 15 ml test tubes. After being rinsed in distilled water
three times, samples were soaked in 10 % Potassium hydroxide (KOH) and
heated for 5 minutes to remove “Humic acids”. Then, samples were
washed by distilled water and treated by Acetic Acid, Glacial, to remove
water. Acetolysis treatment was used to cellulose from samples. Residues
were kept in distilled water for further Scanning Electron Microscopy
(SEM) preparation and immersed in silicon oil for Light Microscopy (LM)
after previous process. I did not use Hydrochloric acid (HCl) in the
pollen extraction because sediments in Clear Lake lack calcium carbonate
components. All chemical treatments were operated in the Acid Lab at
Integrative Biology at University of California, Berkeley.
Samples for the Scanning Electric Microscope (SEM) imaging were prepared
in the Electron Microscope Lab at UC Berkeley. Clean samples, which were
treated by chemical process mentioned above, were done fixation by 2%
glutaraldehyde in 0.1 M Sodium cacodylate buffer, pH 7.2 and by 1%
Osmium tetroxide in 0.1 M sodium cacodylate buffer, pH 7.2 for Post-Fix.
Dehydrate was applied with 35%, 50%, 70%, 80%, 95%, and 100% EOTH.
After Fixation, samples were applied Critical Point Dry. Dried samples
were mounted on stubs using carbon dots and sputter coated for 20 to 25
nanometers thickness with Au. SEM imaging was applied by the machine
model Hitachi S3200N SEM-EDS at University of Exeter and the model
Hitachi S-5000 SEM at University of California, Berkeley. The
magnification 3000X, 10000X, and 15000X were used to take images. Three
images per pollen grain and 2 to 3 grains per species were taken for 23
modern oak species. Total of 190 SEM images of modern oak pollen were
taken and 60 images were used to categorize surface micro-elements.
Samples for Light Microscope (LM) measurements were stored in silicon
oil. Twenty to sixty grains per species were imaged by digital camera in
polar views and equatorial views under LM with magnification 1000X.
The pollen wall consists of three primary layers: the ektexine, endexine
and intine. The ektexine and endexine are the outer sporoponeninous,
acetolysis-resistant layers of the wall preserved in dispersed fossil
pollen. The intine, a non-acetolysis resistant layer, is not preserved
after the chemical treatment and not considered here. The ektexine in
most angiosperm pollen grains comprises small radial, rod-like elements.
The development and distribution of that depend on the extreme
morphologic variability of the exine (Faegri and Iversen, 1964. The
patterns of surface elements of oak pollen vary species-by-species. Most
of them display rod-like elements on surface under scanning electron
microscopes. The terminology of shape and apertures used in this chapter
is followed Erdtman (1952), Faegri and Iversen (1964), Punt et al.
(2007), and Hesse et al. (2009). “Perforate” refers to a surface with
pores much smaller than 1μm in diameter. Microchannels through the
tectum cause those pores (Rowley and Gabarayeva, 2004).
The light microscope (LM) measurements, applied on 710 pollen specimens
(digital images), include one size measurement, polar axis and
equatorial axis, and two shape measurements, P/E ratio and p/Dratio. The scanning electron microscope (SEM) measurements, applied on
22 Quercus modern species native to California and 2 modern
relatives of Quercus lineage. Morphological features of 23 oak
species were divided into evergreen/deciduous, and 3 Quercussections to be statistically analyzed for differentiation patterns. Each
measurement was manually checked by 3 different people before we ran the
statistic programs.
Statistics
I used Image J to measure lengths of LM pollen images. Measurement
results were compared and done T-test under R statistical program,
Rattle GUI (Rattle package). The phylogenetic signals were tested by
Moran’s I test in ape and phytools packages (other used packages:
picante, adephylo, ade4, phylobase, and geiger packages). I used the
phylogenetic trees from Hubert (2014) because it’s the most completed
tree with lengths and taxa included in this study. The graphs were
created by Rattle and ggplot (Rattle and ggplot packages). Rattle GUI is
an open source software which is used for data mining and presents
statistical and visual summaries of data (rattle.togaware.com). It
transforms data to the easily used and modelled graphs (Williams, 2011).
Ape package is featured for phylogenetic framework, ancestral character
analyses, analyses of diversification and macroevolution, computing
distances from DNA sequences. Functions of Ape include generalizing
skyline plots, estimating absolute evolutionary rates and clock-like
trees, translating DNA into AA sequences, accessing sequence alignments,
Moran’s I autocorrelation Index, Bomberg et al.’s correlation structure,
and dating trees with non-contemporaneous sequences (Paradis and
Schliep, 2018). Phytools is a multifunctional phylogenetics package run
under R program. The featured functions of phytools include generating
all possible bi- and multifurcating trees for a set of taxa, performing
ancestral character estimation with a trend using likelihood, fitting a
multiple rate Brownian evolution model using likehood, creating an
animation of Brownian motion evolution with speciation, estimating the
lineage density at each node of the tree, fitting multiple evolutionary
variance-covariance matrices to the tree, conducting y-test, and mapping
temporal intervals on a phylogenetic tree (Revell, 2012).
Table 2 Quercus species investigation for this study